China Deploys Generative AI in Military Intelligence Operations
China's military is deploying generative AI models developed by the National University of Defense Technology to automate intelligence analysis across multiple collection disciplines. The models, including a tool called ChatBIT based on adapted Meta LLaMA, aim to process large data volumes and forecast adversary actions with reduced human involvement.
What happened
China National University of Defense Technology (NUDT), a major defense technology centre, is developing and deploying large language model platforms for military intelligence operations. The models process data from open-source intelligence (OSINT), signals intelligence (SIGINT), geospatial intelligence (GEOINT) and human intelligence (HUMINT).
The Chinese Armed Forces are integrating these models into command-and-analysis units to automate data processing, classify information sources, generate operational reports and forecast adversary actions. The specialized ChatBIT tool, based on adapted Meta LLaMA models, is being tested for intelligence analysis and vulnerability identification.
Reported impact
According to Ukraine's Centre for Countering Disinformation, the automation aims to create a 24/7 digital analyst that reduces the time from information receipt to decision-making by automating previously manual analytical tasks.
Index analysis
The deployment represents integration of autonomous AI systems into military decision-making processes without specified human oversight mechanisms at each stage of analysis or recommendation.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.